Fine-Grained Content Zone Prediction in German Argumentative Essays Using LLMs
Year: 2026
Type: other
Abstract: We introduce FDE-Arg, a new dataset of argumentative student essays in German. We use two Llama models of different sizes to label sentence-level content zones both in FDE-Arg and in an existing dataset of source-dependent argumentative essays. We investigate three approaches for improving model performance: a) Incorporating targeted task information into the prompt text; b) few-shot prompting with up to 10 examples selected on the basis of similarity with the target instance; and c) parameter-efficient finetuning. While the finetuning approach has not been shown to be successful, both incorporating additional information in the prompts and similarity-based few-shot prompting have produced highly promising performance gains compared to the baseline.
Source: Underline Science Inc.
DOI: 10.48448/93zz-ve33
Institution: University of Potsdam
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7170092213
